Every argument clarity score on this site is built from rows on this page. Each
question and answer was assessed with names hidden, the host's own answers included, on
four things from 1 to 5:
directness (does it answer the question asked), coherence (do the ideas follow),
precision (concrete details and clear references), compression (says a lot per word). The weighted
mix (30/30/25/15) is the exchange score. A person's published score averages their exchange
scores on raw tape only, at least 8 of them, shrunk toward the cohort mean.
Full method →
Answered raw tape
D 3 · C 3 · P 4 · Cm 3 3.25
Q world to basically Ignore everything else in every other data source that even could be put in and just operate purely on the order book data, um, just better than everyone else. But so that feels like, okay, I wouldn't know how to, you know, create any extra value there with something else, but it sounds like you found something that they all need in common. Like, what is that?
A Yeah, I think there's, it's, it's a, it's a series of tools across Wall Street. I think it's just one thing. It's just a myopia that, you know, and just sort of a laziness that's enveloped the bigger Companies, again, you know, there's the, so, so we have this data that shows that around five percent of Jane Street and Two Sigma use Bloomberg, and the reason is because they view it as an entry-level tool. You know, uh, Chamath is looking at it and says, wow, this is like an exclusive social network with the best finance, you know, it's true, Peter Thiel's on there, 24 seven, I see his little green light next to his name, but the, uh, the point is that, you know, Wall Street changes, and the tool, the tool sets are changing, even for fundamental equity guys, so Credit card data, you know, to the minute is something.
AI assessment note: “the tool sets are changing, even for fundamental equity guys, so Credit card data”
Answered raw tape
D 3 · C 3 · P 4 · Cm 3 3.25
Q So my, when you say quant, uh, are, is there a bifurcation between like high frequency trading and just quantitative trading?
A Yeah, and I think it's gonna get, my goal actually is to make it more of a spectrum. So the fund I was at yesterday, I said that you guys can do what Renaissance does. You know, it's not a secret anymore. It might have been a secret in 1995, but the amount of kids that have come and left every one of these firms, hopping from Jane Street to Citadel to the next shop, everybody knows what everyone's doing. It's just a question of your risk tolerance, your leverage. Execution is very important, but I think that, you know, trading You know, the, the stock pickers are kind of going the way of dinosaur. And I think by moving up the power curve for Bloomberg, helping people become quants, you know, the, the quant industry has sold, I think this tremendous lie. And again, these are customers. Some, you know, I love those guys, but I think that is in their vested interest to tell people that look at this blackboard with all these math equations. There's no way you guys could understand this. You're too stupid. You have to be an IMO winner. You couldn't possibly come here And make billions, but we saw what Jane Street did.
AI assessment note: “Yeah, and I think it's gonna get, my goal actually is to make it”
Redirected raw tape
D 2 · C 3 · P 2 · Cm 2 2.30
Q Yeah. That's kind of what I was getting at. I'm one of my, Very first questions around the, the, the new terminal would just be ingesting and classifying all this data and then just immediately taking action on it without necessarily having a human in the loop, right?
A Yeah, this is one of the things I want to bring up to my, my, uh, uh, partner's colleagues next time I head out, uh, west is that, you know, one of the fantasies of AGI is that, well, if you do have the machine God, why not unleash it on the stock market? And, you know, it can self-fund you. It can make, you know, a hundred billion dollars. And, you know, you could, you know, Jane treated me twenty billion dollars and nobody would have noticed, you know, last year in profits. So if you have the machine god, and that's, you know, that's basically, I hate to say this, but that's a bunch of old algorithms. Uh, that, that, you know, they've dressed around some, some IMO dressing on it. And so, you know, the real machine God.
AI assessment note: “one of the fantasies of AGI is that, well, if you do have”
Not addressed raw tape
D 1 · C 2 · P 3 · Cm 2 1.95
Q and win the AI researcher race, we should ban high frequency trading. It feels like we might not even need to have that conversation because Mark Zuckerberg is willing to pay as much as Jane Street now. Um, but what is your take on whether or not, uh, economic value or American values are created through the process of high frequency trading? Should we ban it? Is there an answer?
A So two awesome, like quick and funny stories. The first is Citadel published a paper on back in the V 100 days. There's V V 100, A 100, H 100, and then V 100. So in the V 100 days, Citadel found a way to do mat moles faster than NVIDIA did. And it was like the most incredible, you know, find ever. And it's like, you know, how is that possible? And the, and the paper's fascinating because the techniques they used were just remarkable. The second story is, um, so, you know, they're brilliant people, obviously at these firms. Uh, the second story is I haven't hard launched this yet, but you know, um, I'm having a baby with a woman. She's my new partner.
AI assessment note: “So two awesome, like quick and funny stories.”
